Non-contact infrared temperature measurement security check system and method for new energy automobile chassis

Abstract

The invention relates to a non-contact infrared temperature measurement security check system and method for a new energy automobile chassis, and the system comprises at least one license plate recognition camera, at least two geomagnetic sensors, a set of chassis infrared scanning array, and a security check operation platform. The chassis infrared scanning array comprises a waterproof shell and a plurality of infrared card type cameras arranged in the waterproof shell, the security check operation table comprises a geomagnetic signal receiver, a computing workstation and a display screen, and non-contact infrared temperature measurement scanning is conducted on a chassis of the new energy automobile through the infrared card type cameras before the new energy automobile boarding. A corresponding infrared image video stream is generated, a computing workstation equipped with parallel computing acceleration hardware adopts the technologies of image splicing, infrared image recognition and the like to automatically recognize a local temperature overheating area of a chassis and give an alarm when the local temperature overheating area exceeds a temperature threshold value, and therefore potential problems and hidden dangers of motor overheating, battery overheating, line aging, abrasion, short-circuit overheating and the like of a new energy automobile are effectively detected.

Description

Technical Field

[0001] The present invention relates to the technical field of safety detection of new energy vehicles, and in particular to a non-contact infrared temperature measurement safety inspection system and method for a new energy vehicle chassis, which is suitable for chassis safety detection of a new energy vehicle before boarding a roll-on / roll-off ferry across the sea or across a river. Background Art

[0002] In recent years, the country has vigorously developed the new energy vehicle industry, and the number and market share of new energy vehicles have increased year by year. However, thermal runaway fires of new energy vehicle power batteries occur from time to time. Such fires are characterized by sudden fire, rapid fire, rapid spread, and easy re-ignition. In some areas where bridges and tunnels have not yet been built, ferries are the only means of transportation for vehicles to cross the sea and rivers. The sailing time of ferries ranges from more than ten minutes to several hours. Once a new energy vehicle fire occurs during the voyage of a ferry, emergency response is extremely difficult due to the relatively narrow space on the ship and limited rescue equipment and resources.

[0003] The new energy vehicles used by households to cross the sea and river by ferry have great differences in brand, age, condition, battery SOC, etc. The vehicles may have experienced long-distance high-speed driving before boarding the ship, which may lead to motor overheating and battery overheating. There may also be safety hazards such as unexposed battery packs being damaged by collisions, aging and wear of lines, and short circuits. Practice has shown that almost all electrical and mechanical equipment will generate heat before a failure occurs, and infrared thermal imaging temperature measurement technology, as an effective detection method in the field of preventive maintenance, can detect potential faults in a timely, rapid and accurate manner before the fault develops into a serious problem.

[0004] At present, when ferries carry new energy vehicles, the main detection methods used are visible light chassis scanning systems and manual visual security checks using special chassis reflective mirrors. Visible light chassis scanning systems and manual visual security checks can only detect whether the chassis contains contraband and whether there is obvious damage. They cannot detect abnormal temperature conditions in the chassis area or potential faults. They are difficult to meet the needs of comprehensive security checks before new energy vehicles board the ship, and there are major security loopholes. Summary of the invention

[0005] In order to solve the problem that existing detection means cannot detect the abnormal temperature of the chassis of new energy vehicles, resulting in security loopholes, the present invention proposes a non-contact infrared temperature measurement security inspection system for the chassis of new energy vehicles, which performs non-contact temperature scanning on the chassis of the new energy vehicle before boarding, automatically identifies the local overheating area of ​​the chassis and alarms, thereby effectively detecting the hidden dangers of new energy vehicle motor overheating, battery overheating, line aging wear short circuit overheating and other problems. The present invention also relates to a non-contact infrared temperature measurement security inspection method for the chassis of new energy vehicles.

[0006] The technical solution of the present invention is as follows:

[0007] A non-contact infrared temperature measurement security inspection system for a new energy vehicle chassis is used to perform safety inspection on the chassis of a new energy vehicle in a security inspection area, and is characterized by comprising:

[0008] At least one license plate recognition camera, arranged above or on the side of the entrance to the security inspection area, for capturing the license plate number of a new energy vehicle entering the security inspection area and transmitting the license plate number to a computing workstation;

[0009] At least two geomagnetic sensors are respectively arranged on the ground at the entrance and exit of the security inspection area, and are used to sense the new energy vehicle entering and leaving the security inspection area, and generate corresponding geomagnetic trigger signals;

[0010] A chassis infrared scanning array is embedded in the ground of the security inspection area and located between the geomagnetic sensor on the ground at the entrance and the geomagnetic sensor on the ground at the exit of the security inspection area. It includes a waterproof housing and a plurality of infrared card-type cameras arranged in the waterproof housing. Each of the infrared card-type cameras is arranged in one or more rows in a direction perpendicular to the driving direction of the new energy vehicle. An infrared penetration window is arranged on the side of the waterproof housing facing the ground. The infrared card-type camera performs non-contact infrared temperature measurement scanning on the chassis of the new energy vehicle through the infrared penetration window, generates a corresponding infrared image video stream and transmits it to the computing workstation;

[0011] A security inspection console, located near the security inspection area, includes a geomagnetic signal receiver, a computing workstation and a display screen.

[0012] The geomagnetic signal receiver is used to receive the geomagnetic trigger signal of the geomagnetic sensor and transmit it to the computing workstation; the computing workstation is equipped with parallel computing acceleration hardware, which is used to receive the geomagnetic trigger signal sent by the geomagnetic signal receiver; obtain the license plate number from the license plate recognition camera; control the start and stop of the temperature measurement scanning operation of the infrared card-type camera in the chassis infrared scanning array; receive the infrared image video stream output by each infrared card-type camera, and synchronously extract a single frame image from multiple infrared image video streams and then use image stitching technology to synthesize the entire chassis infrared temperature image to reconstruct the thermal distribution map and display it on the display screen, use infrared image recognition technology to automatically identify the temperature abnormal area in the thermal distribution map, trigger an alarm when the local temperature exceeds a preset threshold, and generate discrimination data including abnormal area information and alarm status, and use the license plate number combined with UTC timestamp as the unique identifier of the chassis infrared temperature image and discrimination data storage.

[0013] Preferably, the chassis infrared scanning array is installed by being embedded in the ground of the security inspection area, with its upper surface flush with the ground. It also includes an Ethernet switch arranged in a waterproof housing. The infrared card-type cameras are deployed in multiple rows in a direction vertical to the driving direction of the new energy vehicle and are all connected to the Ethernet switch. The imaging areas of two adjacent infrared card-type cameras in each row overlap to ensure the complete capture of the infrared image. The infrared penetration window adopts a germanium glass window. The size and number of the germanium glass window match the multiple rows of infrared card-type cameras. The infrared card-type cameras generate corresponding infrared image video streams that are transmitted to the computing workstation through the Ethernet switch.

[0014] Preferably, the computing workstation synchronously extracts single-frame images from multiple infrared image video streams, and then uses image stitching technology based on image distortion correction, image enhancement, image feature extraction, image feature tracking, and image feature matching to stitch these single-frame images into a complete chassis infrared temperature image.

[0015] Preferably, the computing workstation adopts infrared image recognition technology based on machine learning, and automatically identifies abnormal temperature gradient areas in the thermal distribution map through a pre-trained convolutional neural network. When the local temperature exceeds the threshold determined according to the new energy vehicle industry standards and historical data statistical analysis, an alarm is triggered through an audible and visual alarm to generate discrimination data including abnormal area information and alarm status.

[0016] Preferably, the security inspection console also includes a data storage device, and the computing workstation uses the license plate number combined with the UTC timestamp as the unique identifier for the chassis infrared temperature image and discrimination data storage, and stores the chassis infrared temperature image and discrimination data in the data storage device to facilitate subsequent data query, management and tracing.

[0017] Preferably, the display screen is integrated with a human-computer interaction interface for: displaying thermal distribution diagrams and alarm information; receiving control instructions from operators, including starting or stopping scanning, adjusting thresholds, and querying historical data; and providing real-time feedback on system status, including scanning progress and equipment operating status.

[0018] A non-contact infrared temperature measurement security inspection method for a new energy vehicle chassis, characterized in that it comprises the following steps:

[0019] Vehicle entry sensing and license plate number acquisition steps: a geomagnetic sensor arranged on the ground at the entrance of the security inspection area senses the new energy vehicle entering the security inspection area, generates a corresponding geomagnetic trigger signal, and a geomagnetic signal receiver integrated in a security inspection operation table near the security inspection area receives the geomagnetic trigger signal and sends it to a computing workstation in the security inspection operation table; at the same time, a license plate recognition camera arranged above or on the side of the entrance of the security inspection area captures the license plate number of the new energy vehicle, and the computing workstation obtains the license plate number from the license plate recognition camera;

[0020] Chassis infrared temperature measurement scanning start step: after the computing workstation receives the geomagnetic trigger signal of the new energy vehicle entering, all infrared card-type cameras in the chassis infrared scanning array installed on the ground are synchronously started to start temperature measurement and video recording; the chassis infrared scanning array includes a waterproof shell and a plurality of infrared card-type cameras arranged in the waterproof shell, each of the infrared card-type cameras is arranged in one or more rows in the vertical direction of the new energy vehicle's driving direction, and an infrared penetration window is arranged on the side of the waterproof shell facing the ground; after starting, the infrared card-type camera performs non-contact infrared temperature measurement and video recording on the chassis of the new energy vehicle through the infrared penetration window, generates a corresponding infrared image video stream and transmits it to the computing workstation; the computing workstation simultaneously receives the infrared image video stream output by all infrared card-type cameras;

[0021] Vehicle exit sensing and scanning stop step: a geomagnetic sensor disposed on the ground at the exit of the security inspection area senses that the new energy vehicle has exited the security inspection area, and generates a corresponding geomagnetic trigger signal. The geomagnetic signal receiver sends the geomagnetic trigger signal to the computing workstation, and the computing workstation stops receiving the infrared image video stream output by all infrared card cameras, and stops the temperature measurement and recording work of all infrared card cameras in the chassis infrared scanning array;

[0022] Thermal distribution map reconstruction step: The computing workstation synchronously extracts single-frame images from multiple infrared image video streams received and then uses image stitching technology to synthesize the entire chassis infrared temperature image to reconstruct the thermal distribution map and display it on the display screen of the security inspection console;

[0023] Abnormal area identification and alarm triggering steps: The computing workstation uses infrared image recognition technology to automatically identify abnormal temperature gradient areas in the thermal distribution map, trigger an alarm when the local temperature exceeds the threshold, and generate identification data containing abnormal area information and alarm status;

[0024] Data storage identification setting steps: The computing workstation uses the license plate number combined with the UTC timestamp as the unique identification for the chassis infrared temperature image and discrimination data storage.

[0025] Preferably, in the chassis infrared temperature measurement scanning start step, the imaging areas of two adjacent infrared card-type cameras in each row overlap to ensure complete coverage and continuous imaging of the new energy vehicle chassis.

[0026] Preferably, in the thermal distribution map reconstruction step, the computing workstation synchronously extracts single-frame images from multiple infrared image video streams, and then uses image stitching technology based on image distortion correction, image enhancement, image feature extraction, image feature tracking and image feature matching to stitch these single-frame images into a complete chassis infrared temperature image.

[0027] Preferably, in the abnormal area identification and alarm triggering step, the computing workstation adopts infrared image recognition technology based on deep learning, and automatically identifies abnormal temperature gradient areas in the thermal distribution map through a pre-trained convolutional neural network. When the local temperature exceeds the threshold determined according to the new energy vehicle industry standards and historical data statistical analysis, an alarm is triggered by an audible and visual alarm, and discrimination data including abnormal area information and alarm status is generated; the computing workstation also performs historical trend analysis on the entire chassis infrared temperature image and the discrimination data to predict potential safety hazards.

[0028] The technical effects of the present invention are as follows:

[0029] The invention relates to a non-contact infrared temperature measurement security inspection system for a chassis of a new energy vehicle. A chassis infrared scanning array is provided, and the system is embedded in the ground. The infrared card-type cameras in a waterproof housing are arranged in one or more rows along the vertical direction of the driving direction of the new energy vehicle. An infrared penetration window is provided on the side of the waterproof housing facing the ground. The infrared card-type camera performs non-contact infrared temperature measurement scanning on the chassis of the new energy vehicle through the infrared penetration window, generates a corresponding infrared image video stream and transmits it to a computing workstation. The computing workstation equipped with parallel computing acceleration hardware adopts image splicing technology, infrared image recognition technology and other technologies to automatically identify the local temperature overheating area of ​​the chassis, and alarms when the temperature exceeds the threshold value, thereby effectively detecting the hidden dangers of the new energy vehicle motor overheating, battery overheating, line aging wear short circuit overheating and other problems. Based on the non-contact infrared temperature measurement technology, potential safety risks are discovered in advance, and the new energy vehicle with chassis hidden dangers and faults is prevented from boarding the ferry roll-on / roll-off ship, which significantly reduces the possibility of fire during the voyage of the ferry, and ensures the safety of life and property. The present invention is also called a new energy vehicle chassis security inspection system based on non-contact infrared temperature measurement. It is specially designed for the application scenario of ferry roll-on / roll-off ships carrying new energy vehicles across the sea or river. It fully considers the special needs and environmental characteristics of new energy vehicle security inspection in this scenario. It can quickly and efficiently complete the security inspection of the new energy vehicle chassis before boarding at the dock, without affecting the normal boarding process of the vehicle, thereby improving the safety and efficiency of ferry operations.

[0030] The present invention directly obtains the license plate number through the license plate recognition camera and transmits it to the computing workstation. Combined with the vehicle sensing function of the geomagnetic sensor, the system can accurately control the detection process, ensure that the detection data of each vehicle is accurately corresponding, and improve the safety and reliability of the entire security inspection system. The chassis infrared scanning array uses a number of infrared card-type cameras with small resolution to form an array, and adopts non-contact infrared temperature measurement technology. The chassis of the new energy vehicle is scanned in real time by the infrared card-type camera to generate an infrared image video stream. This non-contact detection method avoids direct contact with the vehicle chassis, which not only improves the detection efficiency, but also reduces the damage or error that may be caused by contact. It is particularly suitable for vehicle detection scenarios. And because the chassis of new energy vehicles is relatively low, this design can cover a larger scanning area to ensure comprehensive detection of all parts of the chassis. At the same time, compared with conventional large-sized high-resolution infrared cameras, the array composed of infrared card-type cameras is more suitable for the application scenario of close-range infrared temperature measurement of the vehicle chassis, and also realizes the lightweight and miniaturization of the product, which is more conducive to embedded ground installation and construction, reduces the difficulty of system installation and maintenance, and improves the stability and reliability of the system.

[0031] The system is equipped with a high-performance computing workstation and parallel computing acceleration hardware (such as GPU, NPU, TPU, etc.), which can quickly process infrared image video streams, and reconstruct the thermal distribution map by synthesizing the entire chassis infrared temperature image through synchronous extraction of single-frame images and image stitching technology. Combined with advanced infrared image recognition technology, the system can automatically identify abnormal temperature areas to ensure high accuracy and high reliability of detection. When the local temperature exceeds the threshold determined according to the new energy vehicle industry standards and historical data statistical analysis, the system will trigger an alarm in time through sound and light alarms to remind security personnel to pay attention to abnormal temperature areas. At the same time, the system uses the license plate number and UTC timestamp as the unique identifier of data storage, and stores the chassis infrared temperature image and discrimination data in the local data storage device for subsequent data query, management and traceability. Through the analysis of a large amount of historical data, the occurrence law of new energy vehicle chassis failures can be summarized, providing strong data support for the design improvement, maintenance and security inspection standards of new energy vehicles, and promoting the healthy development of the new energy vehicle industry. The system design of the present invention is highly flexible and scalable. The parallel computing acceleration hardware equipped in the computing workstation not only supports current high-performance computing needs, but can also adapt to new technologies that may emerge in the future (such as new parallel computing chips), ensuring the long-term applicability and technological leadership of the system.

[0032] The present invention also relates to a non-contact infrared temperature measurement security inspection method for a new energy vehicle chassis. The method corresponds to the above-mentioned non-contact infrared temperature measurement security inspection system for a new energy vehicle chassis, and can be understood as a method for realizing a non-contact infrared temperature measurement security inspection system for a new energy vehicle chassis, or a non-contact infrared temperature measurement security inspection method for a new energy vehicle chassis based on a non-contact infrared temperature measurement security inspection system for a new energy vehicle chassis. The method includes a vehicle entry sensing and license plate number acquisition step, a chassis infrared temperature measurement scanning start step, a vehicle exit sensing and scanning stop step, a thermal distribution map reconstruction step, an abnormal area identification and alarm triggering step, and a data storage identifier setting step. , which is suitable for chassis safety detection of new energy vehicles before boarding the ferry or ro-ro ship to cross the sea or river. It comprehensively uses non-contact infrared thermal imaging, synchronous extraction, image stitching, infrared image recognition and other technologies to perform non-contact temperature scanning on the chassis of the new energy vehicle before boarding the ship to form a thermal distribution map, automatically identify the local temperature overheating area of ​​the chassis, and alarm when the temperature threshold is exceeded, so as to effectively detect the hidden dangers of new energy vehicle motor overheating, battery overheating, line aging wear short circuit overheating and other problems, and avoid new energy vehicles with hidden dangers and faults in the chassis area from boarding the ship, thereby reducing the fire safety risk of ferry ro-ro ships carrying new energy vehicles across the sea or across the river. The present invention integrates advanced non-contact infrared temperature measurement technology, high-performance computing acceleration hardware and intelligent image processing algorithms to provide an efficient, accurate and reliable new energy vehicle chassis security inspection solution, which significantly improves the efficiency and safety of security inspections, and provides a broad space for the development of future technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a structural schematic diagram of the non-contact infrared temperature measurement security inspection system for the chassis of a new energy vehicle of the present invention.

[0034] Figure 2 It is a schematic diagram of the horizontal arrangement of the infrared card-type camera of the present invention.

[0035] Figure 3 This is a working flow chart of the non-contact infrared temperature measurement security inspection system for the chassis of a new energy vehicle of the present invention.

[0036] Examples of the labels in the figure are as follows:

[0037] 1—License plate recognition camera; 2—Waterproof housing; 3—Infrared card camera; 4—Germanium glass window; 5—Ethernet switch; 6—Security inspection console; 61—Geomagnetic signal receiver; 62—Computing workstation; 63—Display screen; 64—Power adapter; 65—Keyboard and mouse; 66—Speaker / buzzer. DETAILED DESCRIPTION

[0038] The present invention will be described below in conjunction with the accompanying drawings.

[0039] Aiming at the application scenario of ferry ro-ro ships carrying new energy vehicles across the sea or river, the present invention relates to a non-contact infrared temperature measurement security inspection system for the chassis of new energy vehicles, which is used to perform safety inspection on the chassis of new energy vehicles in the security inspection area. Figure 1 As shown, it includes at least one license plate recognition camera 1, at least two geomagnetic sensors (referred to as geomagnetic, such as Figure 1 The 1# geomagnetic sensor and 2# geomagnetic sensor shown in the figure), a chassis infrared scanning array and a security inspection console 6. The chassis infrared scanning array includes a waterproof housing 2 and a plurality of infrared card-type cameras 3 and an Ethernet switch 5 all arranged in the waterproof housing 2. The waterproof housing 2 is provided with an infrared penetration window on the side facing the ground. The window material has infrared penetration characteristics. In this embodiment, a germanium glass window 4 is used. The security inspection console 6 integrates components such as a geomagnetic signal receiver 61, a computing workstation 62, a display screen 63, a power adapter 64, a keyboard and mouse 65, and a speaker / buzzer 66. Specifically,

[0040] The license plate recognition camera 1 is arranged above or on the side of the entrance to the security inspection area, and uses optical character recognition (OCR) technology to capture the license plate number of the new energy vehicle entering the security inspection area, and transmits the license plate number to the computing workstation 62.

[0041] Two geomagnetic sensors, 1# geomagnetic sensor is set on the ground at the entrance of the security inspection area, and 2# geomagnetic sensor is set on the ground at the exit of the security inspection area. They are used to sense the entry and exit of new energy vehicles in the security inspection area and generate corresponding geomagnetic trigger signals.

[0042] A chassis infrared scanning array is embedded in the ground of the security inspection area, with its upper surface flush with the ground, and is located between the 1# geomagnetic sensor on the ground at the entrance of the security inspection area and the 2# geomagnetic sensor on the ground at the exit, wherein the waterproof housing 2 is a fully waterproof housing, and each of the infrared card-type cameras 3 is arranged in one or more rows along the vertical direction of the driving direction of the new energy vehicle (or arranged horizontally). This embodiment deploys multiple rows, specifically two rows arranged horizontally, and all the infrared card-type cameras 3 are connected to the Ethernet switch 5. The infrared card-type cameras are arranged horizontally at a spacing such as Figure 2 As shown, it can be ensured that the chassis of the new energy vehicle with the lowest chassis has an overlapping area in the imaging of two adjacent infrared card-type cameras 3 to ensure the complete capture of the infrared image. The size and number of the germanium glass windows 4 match the multiple rows of infrared card-type cameras 3 (such as Figure 1 As shown in the figure, each infrared card-type camera 3 performs a non-contact infrared temperature measurement scan on the chassis of the new energy vehicle through the corresponding germanium glass window 4, generates a corresponding infrared image video stream and transmits it to the computing workstation 62 through the Ethernet switch 5.

[0043] The security inspection operation table 6 is arranged near the security inspection area, wherein the geomagnetic signal receiver 61 is used to receive the geomagnetic trigger signal of the geomagnetic sensor and transmit it to the computing workstation 62; the computing workstation 62 is equipped with parallel computing acceleration hardware, which is not limited to GPU, but can also be NPU processing unit hardware, or TPU tensor processing unit hardware, as well as other parallel computing hardware for image and data processing acceleration spawned by future new technologies; the computing workstation 62 is used to receive the geomagnetic trigger signal sent by the geomagnetic signal receiver 61; obtain the license plate number from the license plate recognition camera 1; control the infrared card camera in the chassis infrared scanning array; The camera 3 starts and stops the temperature measurement scanning operation; receives the infrared image video stream output by each infrared card camera 3, and synchronously extracts a single frame image from multiple infrared image video streams and then uses image stitching technology to synthesize the entire chassis infrared temperature image to reconstruct the thermal distribution map and display it on the display screen 63; uses infrared image recognition technology to automatically identify the temperature abnormal area in the thermal distribution map, triggers an alarm when the local temperature exceeds the preset temperature threshold, and generates discrimination data containing abnormal area information and alarm status, and uses the license plate number combined with the UTC timestamp as the unique identifier of the chassis infrared temperature image and discrimination data storage. The display screen 63 integrates a human-machine interaction interface for displaying the thermal distribution map and alarm information; receives the operator's control instructions, including starting or stopping scanning, adjusting thresholds, querying historical data, etc.; and provides real-time feedback on the system status, including scanning progress, equipment operation status, etc. The power adapter 64 provides a stable power supply for the geomagnetic signal receiver 61 in the security inspection console 6, and the infrared card camera 3 and Ethernet switch 5 in the chassis scanning array, etc., to ensure the normal operation of the system. The keyboard and mouse 65 are input devices for the operator to interact with the system and are used to control the operation of the computing workstation, such as starting or stopping scanning, adjusting thresholds, data query, etc. The speaker / buzzer 66 can issue an audible alarm, and cooperate with the display screen 63 human-machine interface graphics to flash, thereby issuing an alarm prompt.

[0044] The computing workstation 62 has powerful computing capabilities and can meet the needs of complex tasks such as scientific computing, engineering design, data analysis, and graphics rendering. It is equipped with parallel computing acceleration hardware to support high-performance parallel computing. Whether it is large-scale 3D modeling, animation rendering, or deep learning training and reasoning, scientific data simulation, etc., it can play an important role and provide users with efficient computing support. The computing workstation 62 completes the reconstruction of the thermal distribution map, the identification and alarm triggering of abnormal areas, and the setting of data storage identifiers. After synchronously extracting single-frame images (i.e., time point alignment, infrared images taken at the same time) from multiple infrared image video streams (also known as infrared image frame sequences), it is preferred to use image stitching technology based on image distortion correction, image enhancement, image feature extraction, image feature tracking, and image feature matching to stitch these single-frame images together, and then synthesize the entire chassis infrared temperature image to achieve reconstruction of the thermal distribution map. Then, the infrared image recognition technology based on machine learning is used to automatically identify the abnormal temperature gradient area in the thermal distribution map through the pre-trained convolutional neural network. When the local temperature exceeds the threshold determined according to the new energy vehicle industry standards and historical data statistical analysis, the alarm can be triggered by sound and light alarms to generate discrimination data containing abnormal area information and alarm status. It can quickly process infrared image video streams, synthesize the entire chassis infrared temperature image through specific image stitching technology to reconstruct the thermal distribution map, and combine with advanced infrared image recognition technology. The system can automatically identify the temperature abnormal area to ensure high accuracy and high reliability of detection.

[0045] Preferably, the security inspection console 6 also includes a data storage device, and the computing workstation 62 uses the license plate number combined with the UTC timestamp as the unique identifier for the chassis infrared temperature image and discrimination data storage, and stores the chassis infrared temperature image and discrimination data in the data storage device to facilitate subsequent data query, management and tracing.

[0046] Figure 3 This is the workflow diagram of the non-contact infrared temperature measurement security inspection system for the chassis of new energy vehicles in the present invention. Before the new energy vehicle arrives at the dock to board the ship, the infrared temperature measurement security inspection operation steps are as follows:

[0047] ① The new energy vehicle slows down and drives into the security inspection area, passing over the 1# geomagnetic sensor. The geomagnetic signal receiver 61 sends the geomagnetic trigger signal (or trigger result) generated by the 1# geomagnetic sensor to the computing workstation 62. At this time, the computing workstation 62 obtains the license plate number from the license plate recognition camera 1;

[0048] ②, the computing workstation 62 synchronously starts all infrared card-type cameras 3 in the chassis infrared scanning array to start recording;

[0049] ③, the computing workstation 62 simultaneously receives the infrared image video streams output by all infrared card-type cameras 3;

[0050] ④, the new energy vehicle leaves the security inspection area and passes over the 2# geomagnetic sensor. The geomagnetic signal receiver 61 sends the geomagnetic trigger signal (or the trigger result) generated by the 2# geomagnetic sensor to the computing workstation 62. At this time, the computing workstation 62 stops receiving the infrared image video stream output by all infrared card cameras 3;

[0051] ⑤, the computing workstation 62 stops the recording work of all infrared card-type cameras 3 in the chassis infrared scanning array;

[0052] ⑥, the computing workstation 62 uses image distortion correction, image enhancement, image feature extraction, image feature tracking, image feature matching and stitching and other image stitching technologies to stitch together the entire chassis infrared temperature image, thereby reconstructing the thermal distribution map and displaying it on the display screen 63 of the security inspection operation station 6;

[0053] ⑦, the computing workstation 62 automatically identifies the abnormal temperature gradient area in the thermal distribution map, triggers an alarm when the local temperature exceeds the threshold, prompts the security personnel to conduct further in-depth inspection, and generates identification data including abnormal area information and alarm status;

[0054] ⑧, the computing workstation 62 uses the license plate number plus the UTC timestamp as the unique identifier for storing the chassis infrared scanning image and identification data.

[0055] The present invention also relates to a non-contact infrared temperature measurement security inspection method for a new energy vehicle chassis. The method corresponds to the above-mentioned non-contact infrared temperature measurement security inspection system for a new energy vehicle chassis, and can be understood as a method for realizing a non-contact infrared temperature measurement security inspection system for a new energy vehicle chassis. The method includes a vehicle entry sensing and license plate number acquisition step, a chassis infrared temperature measurement scanning start step, a vehicle exit sensing and scanning stop step, a thermal distribution map reconstruction step, an abnormal area identification and alarm triggering step, and a data storage identifier setting step. Specifically, reference can be made to Figure 3 The process shown:

[0056] 1. Vehicle entry sensing and license plate number acquisition steps: A geomagnetic sensor installed on the ground at the entrance of the security inspection area senses the new energy vehicle entering the security inspection area, generates a corresponding geomagnetic trigger signal, and a geomagnetic signal receiver integrated in a security inspection operation table near the security inspection area receives the geomagnetic trigger signal and sends it to a computing workstation in the security inspection operation table; at the same time, a license plate recognition camera installed above or on the side of the entrance of the security inspection area captures the license plate number of the new energy vehicle, and the computing workstation obtains the license plate number from the license plate recognition camera;

[0057] 2. Chassis infrared temperature measurement scanning start step: after the computing workstation receives the geomagnetic trigger signal of the new energy vehicle entering, all infrared card-type cameras in the chassis infrared scanning array installed on the ground are synchronously started to start temperature measurement and video recording; the chassis infrared scanning array includes a waterproof housing and a plurality of infrared card-type cameras arranged in the waterproof housing, each of the infrared card-type cameras is arranged in one or more rows in the vertical direction of the driving direction of the new energy vehicle, and an infrared penetration window is arranged on the side of the waterproof housing facing the ground; after starting, the infrared card-type camera performs non-contact infrared temperature measurement and video recording of the chassis of the new energy vehicle through the infrared penetration window, generates a corresponding infrared image video stream and transmits it to the computing workstation; the computing workstation simultaneously receives the infrared image video stream output by all infrared card-type cameras;

[0058] 3. Vehicle exit sensing and scanning stop step: The geomagnetic sensor set on the ground at the exit of the security inspection area senses that the new energy vehicle has exited the security inspection area, and generates a corresponding geomagnetic trigger signal. The geomagnetic signal receiver sends the geomagnetic trigger signal to the computing workstation, and the computing workstation stops receiving the infrared image video stream output by all infrared card cameras, and stops the temperature measurement and recording work of all infrared card cameras in the chassis infrared scanning array;

[0059] 4. Thermal distribution map reconstruction step: The computing workstation synchronously extracts single-frame images from multiple infrared image video streams received and then uses image stitching technology to synthesize the entire chassis infrared temperature image to reconstruct the thermal distribution map and display it on the display screen of the security inspection console;

[0060] 5. Abnormal area identification and alarm triggering steps: The computing workstation uses infrared image recognition technology to automatically identify abnormal temperature gradient areas in the thermal distribution map, trigger an alarm when the local temperature exceeds the threshold, and generate identification data containing abnormal area information and alarm status;

[0061] 6. Data storage identification setting steps: The computing workstation uses the license plate number combined with the UTC timestamp as the unique identification for the chassis infrared temperature image and discrimination data storage.

[0062] Furthermore, in the chassis infrared temperature measurement scanning start step, the imaging areas of two adjacent infrared card-type cameras in each row overlap to ensure complete coverage and continuous imaging of the new energy vehicle chassis.

[0063] Furthermore, in the thermal distribution map reconstruction step, the computing workstation synchronously extracts single-frame images from multiple infrared image video streams, and then uses image stitching technology based on image distortion correction, image enhancement, image feature extraction, image feature tracking and image feature matching to stitch these single-frame images into a complete chassis infrared temperature image.

[0064] Furthermore, in the abnormal area identification and alarm triggering step, the computing workstation adopts infrared image recognition technology based on deep learning, and automatically identifies abnormal temperature gradient areas in the thermal distribution map through a pre-trained convolutional neural network. When the local temperature exceeds the threshold determined according to the new energy vehicle industry standards and historical data statistical analysis, an alarm is triggered by sound and light alarms to generate discrimination data including abnormal area information and alarm status.

[0065] Furthermore, the computing workstation also performs a historical trend analysis on the entire chassis infrared temperature image and the discrimination data to predict potential safety hazards.

[0066] The present invention is based on a non-contact infrared temperature measurement security inspection system and method for the chassis of new energy vehicles. Aiming at the application scenario of ferry roll-on / roll-off ships carrying new energy vehicles across the sea or river, the present invention provides an efficient, accurate and reliable new energy vehicle chassis security inspection solution by integrating advanced non-contact infrared temperature measurement technology, high-performance computing acceleration hardware and intelligent image processing algorithms (image stitching technology, infrared image recognition and other technologies). The solution can automatically identify the local overheating area of ​​the chassis and alarm when the temperature exceeds the threshold, thereby effectively detecting hidden dangers such as overheating of the new energy vehicle motor, overheating of the battery, aging and wear of the line, short circuit and overheating, thereby improving the security inspection efficiency and safety, preventing new energy vehicles with chassis hidden dangers and faults from boarding the ferry roll-on / roll-off ship, significantly reducing the possibility of fire on the ferry during navigation, and ensuring the safety of life and property.

[0067] It should be noted that the above-described specific implementations can enable those skilled in the art to more fully understand the invention, but do not limit the invention in any way. Therefore, although this specification has described the invention in detail with reference to the drawings and embodiments, those skilled in the art should understand that the invention can still be modified or replaced by equivalents. In short, all technical solutions and improvements that do not deviate from the spirit and scope of the invention should be included in the protection scope of the patent for the invention.

Claims

1. A non-contact infrared temperature measurement security inspection system for new energy vehicle chassis, used for safety inspection of new energy vehicle chassis in the security inspection area, characterized in that: include: At least one license plate recognition camera, arranged above or on the side of the entrance to the security inspection area, for capturing the license plate number of a new energy vehicle entering the security inspection area and transmitting the license plate number to a computing workstation; At least two geomagnetic sensors are respectively arranged on the ground at the entrance and exit of the security inspection area, and are used to sense the new energy vehicle entering and leaving the security inspection area, and generate corresponding geomagnetic trigger signals; A chassis infrared scanning array is embedded in the ground of the security inspection area and located between the geomagnetic sensor on the ground at the entrance and the geomagnetic sensor on the ground at the exit of the security inspection area. It includes a waterproof housing and a plurality of infrared card-type cameras arranged in the waterproof housing. Each of the infrared card-type cameras is arranged in one or more rows in a direction perpendicular to the driving direction of the new energy vehicle. An infrared penetration window is arranged on the side of the waterproof housing facing the ground. The infrared card-type camera performs non-contact infrared temperature measurement scanning on the chassis of the new energy vehicle through the infrared penetration window, generates a corresponding infrared image video stream and transmits it to the computing workstation; A security inspection console, located near the security inspection area, includes a geomagnetic signal receiver, a computing workstation and a display screen. The geomagnetic signal receiver is used to receive the geomagnetic trigger signal of the geomagnetic sensor and transmit it to the computing workstation; The computing workstation is equipped with parallel computing acceleration hardware and is used to receive the geomagnetic trigger signal sent by the geomagnetic signal receiver; The license plate number is obtained from the license plate recognition camera; the start and stop of the temperature measurement scanning operation of the infrared card-type camera in the chassis infrared scanning array is controlled; the infrared image video stream output by each infrared card-type camera is received, and a single frame image is synchronously extracted from multiple infrared image video streams, and then the image stitching technology is used to synthesize the entire chassis infrared temperature image, so as to reconstruct the thermal distribution map and display it on the display screen, and the infrared image recognition technology is used to automatically identify the temperature abnormal area in the thermal distribution map, and an alarm is triggered when the local temperature exceeds a preset threshold, and discrimination data including abnormal area information and alarm status is generated, and the license plate number is combined with the UTC timestamp as the unique identifier for the chassis infrared temperature image and discrimination data storage.

2. The non-contact infrared temperature measurement security inspection system for chassis of new energy vehicles according to claim 1 is characterized in that: The chassis infrared scanning array is installed by embedding in the ground of the security inspection area, and its upper surface is flush with the ground. It also includes an Ethernet switch arranged in a waterproof shell. The infrared card-type cameras are deployed in multiple rows in the vertical direction of the driving direction of the new energy vehicle and are all connected to the Ethernet switch. The imaging areas of two adjacent infrared card-type cameras in each row overlap to ensure the complete capture of the infrared image. The infrared penetration window adopts a germanium glass window. The size and number of the germanium glass window match the multiple rows of infrared card-type cameras. The infrared card-type cameras generate corresponding infrared image video streams that are transmitted to the computing workstation through the Ethernet switch.

3. The non-contact infrared temperature measurement security inspection system for the chassis of a new energy vehicle according to claim 1 or 2, characterized in that: The computing workstation synchronously extracts single-frame images from multiple infrared image video streams, and then uses image stitching technology based on image distortion correction, image enhancement, image feature extraction, image feature tracking, and image feature matching to stitch these single-frame images into a complete chassis infrared temperature image.

4. The non-contact infrared temperature measurement security inspection system for chassis of new energy vehicles according to claim 3 is characterized in that: The computing workstation adopts infrared image recognition technology based on machine learning, and automatically identifies abnormal temperature gradient areas in the thermal distribution map through a pre-trained convolutional neural network. When the local temperature exceeds the threshold determined according to the new energy vehicle industry standards and historical data statistical analysis, an alarm is triggered through an audible and visual alarm to generate discrimination data including abnormal area information and alarm status.

5. The non-contact infrared temperature measurement security inspection system for chassis of new energy vehicles according to claim 1 is characterized in that: The security inspection console also includes a data storage device. The computing workstation uses the license plate number combined with the UTC timestamp as the unique identifier for storing the chassis infrared temperature image and the discrimination data, and stores the chassis infrared temperature image and the discrimination data in the data storage device to facilitate subsequent data query, management and tracing.

6. The non-contact infrared temperature measurement security inspection system for chassis of new energy vehicles according to claim 1 is characterized in that: The display screen integrates a human-machine interactive interface for: displaying thermal distribution diagrams and alarm information; receiving control instructions from operators, including starting or stopping scanning, adjusting thresholds, and querying historical data; and providing real-time feedback on system status, including scanning progress and equipment operating status.

7. A non-contact infrared temperature measurement security inspection method for chassis of new energy vehicles, characterized in that: The steps include: Vehicle entry sensing and license plate number acquisition steps: a geomagnetic sensor arranged on the ground at the entrance of the security inspection area senses the new energy vehicle entering the security inspection area, generates a corresponding geomagnetic trigger signal, and a geomagnetic signal receiver integrated in a security inspection operation table near the security inspection area receives the geomagnetic trigger signal and sends it to a computing workstation in the security inspection operation table; at the same time, a license plate recognition camera arranged above or on the side of the entrance of the security inspection area captures the license plate number of the new energy vehicle, and the computing workstation obtains the license plate number from the license plate recognition camera; Chassis infrared temperature measurement scanning start step: after the computing workstation receives the geomagnetic trigger signal of the new energy vehicle entering, all infrared card-type cameras in the chassis infrared scanning array installed on the ground are synchronously started to start temperature measurement and video recording; the chassis infrared scanning array includes a waterproof shell and a plurality of infrared card-type cameras arranged in the waterproof shell, each of the infrared card-type cameras is arranged in one or more rows in the vertical direction of the new energy vehicle's driving direction, and an infrared penetration window is arranged on the side of the waterproof shell facing the ground; after starting, the infrared card-type camera performs non-contact infrared temperature measurement and video recording on the chassis of the new energy vehicle through the infrared penetration window, generates a corresponding infrared image video stream and transmits it to the computing workstation; the computing workstation simultaneously receives the infrared image video stream output by all infrared card-type cameras; Vehicle exit sensing and scanning stop step: a geomagnetic sensor disposed on the ground at the exit of the security inspection area senses that the new energy vehicle has exited the security inspection area, and generates a corresponding geomagnetic trigger signal. The geomagnetic signal receiver sends the geomagnetic trigger signal to the computing workstation, and the computing workstation stops receiving the infrared image video stream output by all infrared card cameras, and stops the temperature measurement and recording work of all infrared card cameras in the chassis infrared scanning array; Thermal distribution map reconstruction step: The computing workstation synchronously extracts single-frame images from multiple infrared image video streams received and then uses image stitching technology to synthesize the entire chassis infrared temperature image to reconstruct the thermal distribution map and display it on the display screen of the security inspection console; Abnormal area identification and alarm triggering steps: The computing workstation uses infrared image recognition technology to automatically identify abnormal temperature gradient areas in the thermal distribution map, trigger an alarm when the local temperature exceeds the threshold, and generate identification data containing abnormal area information and alarm status; Data storage identification setting steps: The computing workstation uses the license plate number combined with the UTC timestamp as the unique identification for the chassis infrared temperature image and discrimination data storage.

8. The non-contact infrared temperature measurement safety inspection method for the chassis of a new energy vehicle according to claim 7 is characterized in that: In the chassis infrared temperature measurement scanning start step, the imaging areas of two adjacent infrared card-type cameras in each row overlap to ensure complete coverage and continuous imaging of the new energy vehicle chassis.

9. The non-contact infrared temperature measurement safety inspection method for the chassis of a new energy vehicle according to claim 7 or 8, characterized in that: In the thermal distribution map reconstruction step, the computing workstation synchronously extracts single-frame images from multiple infrared image video streams, and then uses image stitching technology based on image distortion correction, image enhancement, image feature extraction, image feature tracking and image feature matching to stitch these single-frame images into a complete chassis infrared temperature image.

10. The non-contact infrared temperature measurement safety inspection method for chassis of new energy vehicles according to claim 9 is characterized in that: In the abnormal area identification and alarm triggering step, the computing workstation adopts infrared image recognition technology based on deep learning, and automatically identifies abnormal temperature gradient areas in the thermal distribution map through a pre-trained convolutional neural network. When the local temperature exceeds the threshold determined according to the new energy vehicle industry standards and historical data statistical analysis, an alarm is triggered through an audible and visual alarm to generate discrimination data including abnormal area information and alarm status; the computing workstation also performs historical trend analysis on the entire chassis infrared temperature image and the discrimination data to predict potential safety hazards.

Images